Qwen: Qwen3.8 27B vs OpenAI: GPT-5.6 Terra

For a sample job of 1 million input tokens and 250,000 output tokens, Qwen3.8 27B costs $1.17 and GPT-5.6 Terra costs $5.00, so Qwen3.8 27B is about 4.3 times cheaper for the same work.

Both take in a similar amount of text: 1,000,000 and 1,050,000 tokens, about 1,500 and 1,575 pages.

Only Qwen3.8 27B offers video input.

On our blended benchmark score GPT-5.6 Terra leads 71.1 to 55.6.

Across the 20 tasks where they differ, Qwen3.8 27B scores higher on 7 (Code Completion, Code Documentation, Regex Writing and others), GPT-5.6 Terra on 13 (SQL Generation, Code Review, Code Refactoring and others).

Qwen3.8 27B was released in August 2026 and GPT-5.6 Terra in July 2026.

Pick Qwen3.8 27B if you need lower cost or video input; pick GPT-5.6 Terra if you need benchmark performance.

Costs use current list prices per million tokens; page counts assume about 0.75 words per token and 500 words per page.